Calibration Approach Of NO₂ Electrochemical Sensor By Cavity Ring-Down Technique

As an important trace gas in the atmosphere,NO2 plays an important role in the formation of ozone;high concentrations of NO2 not only causes acid rain and photochemical smog,restrict the sustainable development of the economy,but also damage the ecological environment and endanger human health.NO2 monitoring instruments are mainly based on chemiluminescence and spectroscopy,the cost of these instruments restricts their wide application in dense monitoring networks,and it is difficult to capture changes in gas concentration on a fine spatial-temporal resolution scale.In recent years,electrochemical gas sensors have been widely used in the field of atmospheric environmental NO2 monitoring due to their advantages of low cost,low power consumption,high time resolution,flexibility and easy operation.However,in practical use,the NO2 electrochemical sensor is directly exposed to complex and dynamic environmental conditions,and it is easy to drift due to environmental factors,which affects the measurement accuracy.In this paper,the calibration approach of NO2 electrochemical sensor by cavity ring-down technique was studied,the NO2 cavity ring-down absorption spectroscopy system is used as the reference standard to calibrate the NO2 electrochemical sensor(NO2-B43F),the sensor calibration algorithm models were established according to different environmental humidity conditions in field experiment,which solves the problem of sensor’s zero drift and improves the measurement accuracy.Specific studies were as follows:1)The electrochemical sensor system was integrated and the performance of the sensor was studied,the linear response of the sensor to NO2 calibration gas was tested.The sensitivity and detection limit of the NO2-B43F sensor were determined to be 0.156mV/ppb and 2.19ppb(3σ),respectively.The sensor’s transient response,gas cross-interference,and temperature and humidity zero point characteristics were studied.In addition to the linear response to the target gas,the working electrode of the NO2-B43F sensor has nonlinear response to humidity.2)A set of NO2 cavity ring-down absorption spectroscopy system was set up,the detect limit of the system is 42ppt(ls),and switching sampling was adopted to eliminate the influence of environmental factors.The comparison test was carried out,in the measurement of NO2 calibration gas,it has good consistency with the commercial NOx analyzer with a slope of 1.003 and R2 of 0.999;and the NO2 in atmospheric environment was monitored by comparing with long path differential optical absorption spectroscopy(LP-DOAS),the R2 is 0.939,which ensured its accuracy and reliability as the reference standard for electrochemical sensors.3)According to different humidity characteristics and sampling conditions,the sensor data calibration model is established by using quadratic regression model and the BP neural network algorithm.When the humidity was not high(RH<60%),the quadratic regression model was used for correction;when the humidity varies widely,the LM training algorithm combined with the BP neural network was adopted.When controlling the sampling flow rate,the 4-11-1 network structure was used;when directly exposed to the environment,multi-sensors data fusion combined with BP neural network algorithm can effectively improve the measurement accuracy,and the R2 between the calibration result and the reference value was 0.71.